How do I know which model to choose?

[godofecht]

I am not sure whether to use LSTM, RNN, or TCN.
Perhaps we can run some performance comparison tests.

[ggkountouras]

Feel free to try them and post benchmarks!

I consider audio quality to be more important than performance, in which case LSTM > WaveNet > TCN. Performance-wise, it's the other way around, but since none of the 3 models are what I would consider "production-ready" in terms of audio quality, there is little point in benchmarking them.